PEST Boston III – September 5th 2007
Yesterday was PEST Boston III, my first PEST!
I arrived in Boston the day before – my first time in America – and although passport control and customs were a bit of a trial (secondary passport control established I knew who I was and who I work for, and the sniffer dog established that I sometimes carry sandwiches or fruit to work in my laptop bag!) I eventually collected my rental car – with satnav – and drove straight into the first tunnel out of Boston Logan International…
Since then I’ve been having a brilliant time – among others, I’ve met with people from Northeastern University and Wellesley College, and tomorrow I’m meeting with Stanley Sclaroff of Boston University, where we are about to commence a new scholarship program – exciting times!
It has been really great to meet some people I had already heard about from previous PESTs, and there were a couple of new faces in the group who everyone enjoyed meeting – Hongwei Xi from Boston University, and Sara-Jane Haven from Tufts Health Plan.
The venue was Owen O’Leary’s in Southborough, which worked really well for the number of people we had – but the suggestion from PEST II of a round table in a more secluded area still stands, and if anyone has any suggestions please let us know!
We had two experience reports in the evening – the first from Nancy O’Leary of HP, on automation, which led to an interesting discussion that could easily have gone on for ages had Don not gently reminded me of my role for the evening J
The second experience report came from Dan Downing of Mentora – a case study on the subject of load and performance testing – a thought-provoking account of working with an organisation who appeared to have made some classic mistakes.
Things we learnt this time:
- Automation tools should have comprehensive evalutaion versions to assess their usefulness.
- Beer can have bits of fruit in it and still taste good (our server was pushing the Plum Beer to get a free T-shirt, which we managed for her!)
- Estimation of the maximum expected number of concurrent users for an application can be calculated in some interesting ways.
- Throwing hardware at a performance problem can easily make things worse not better overall.
- Architecture problems might be obvious, but if they're hard to solve they may stay an 'elephant in the room' until it's too late.
- Academia and industry see some things in different ways and each can learn from the other - it was good to have Hongwei with us at this PEST to give a slightly different perspective!
- It is important to establish a set of metrics for automation tools, so their efficiency can be accurately determined for repeat usage
Our PEST website is nearly ready to go live, so hopefully we will soon have a forum for the continuation of some of the lively discussions we have had, and a place for people to keep in touch between events.
In the meantime if anyone would like more information about PEST (or about our scholarship programs) please send me or Stewart an email.
I hope to meet everyone again at some point in the future – thank you to everyone who attended for making it such a good evening. Remember the point of PEST: none of us is as smart as all of us!
Jon (jonathan.wright@transitionconsulting.co.uk)
P.S. Additional thanks to Don St. Pierre for his helpful advice and for booking the venue!
